{"id":{"repo_id":"msu","oai_identifier":"oai:d.lib.msu.edu:etd_52950"},"canonical_url":"https://search.ndltd.org/etd/msu/oai:d.lib.msu.edu:etd_52950","repository":{"repo_id":"msu","name":"Michigan State University","base_url":"https://d.lib.msu.edu/oai"},"display":{"title":"On-Device Semi-Supervised Learning For Healthcare Applications","abstract":"Thesis (Ph.D.)--Michigan State University. Electrical and Computer Engineering - Doctor of Philosophy, 2026","abstract_html":"Thesis (Ph.D.)--Michigan State University. Electrical and Computer Engineering - Doctor of Philosophy, 2026","abstract_has_math":false,"creators":["Roy, Avirup"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Biswas, Subir","Ren, Jian","Kulkarni, Sandeep","Traganitis, Panagiotis"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026","date_published":"2026","updated_at":"2026-11-18T18:59:34Z","subjects":["Electrical engineering","Computer engineering"],"languages":["eng"],"rights":["Attribution 4.0 International"],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:52950","local:Roy_grad.msu_0128D_21961"],"render_values":[{"text":"etd:52950","href":null,"code":true},{"text":"local:Roy_grad.msu_0128D_21961","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/doi:10.25335/kc65-vd80","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/doi:10.25335/kc65-vd80","record_label":"Repository record","record_kind":"doi","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://d.lib.msu.edu/oai?verb=GetRecord&metadataPrefix=oai_dc&identifier=oai%3Ad.lib.msu.edu%3Aetd_52950","prefix":"oai_dc"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Biswas, Subir","Ren, Jian","Kulkarni, Sandeep","Traganitis, Panagiotis"]},{"key":"dc:creator","label":"Author","values":["Roy, Avirup"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2026"]},{"key":"dc:relation","label":"Dc Relation","values":["Electronic Theses & Dissertations"]},{"key":"dc:type","label":"Dc Type","values":["text","Electronic dissertations"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Electrical engineering","Computer engineering"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Attribution 4.0 International"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:52950","local:Roy_grad.msu_0128D_21961","https://doi.org/doi:10.25335/kc65-vd80"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Thesis (Ph.D.)--Michigan State University. Electrical and Computer Engineering - Doctor of Philosophy, 2026","Mobile health monitoring applications heavily rely on Embedded Cyber-Physical Systems (E-CPS) to provide real-time measurement and diagnostics insights into user health in the areas of physical activity, hydration level, various vital signs, and lifestyle in general. These systems demand not only high accuracy but also rapid system performance in order to ensure timely health interventions. Delays in processing can result in suboptimal outcomes, especially in critical health scenarios where immediate actions are required. To reduce operational latency and improve performance, intelligent E-CPS typically employ pre-trained supervised learning models to infer patterns based on real-time health data. However, the traditional approach to training these models presents challenges, such as limited access to labeled health data, difficulties in developing personalized learning models, and privacy concerns arising from the transmission of sensitive health data to external servers for training.This dissertation introduces a semi-supervised learning (SSL) framework to address such limitations of conventional supervised learning in mobile health monitoring applications. The proposed SSL framework leverages new unsupervised clustering and cluster labeling techniques to learn from sparsely labeled data points, making it particularly useful in scenarios where sufficient labeled data is unavailable. By training directly on data collected from each user’s device, this approach facilitates person-specific learning, enabling the system to adapt to the unique health patterns of individual users. The on-device learning mechanism ensures that all data remains on the user’s device, eliminating the need for data transmission to external servers, thereby enhancing privacy and reducing communication costs. Additionally, the framework is optimized for resource-constrained devices, consuming minimal computational power and battery life, making it ideal for wearable and mobile challenges of real-time health monitoring while ensuring that devices remain responsive and effective for continuous, personalized health tracking.","Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).","Includes bibliographical references."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf","1 online resource (146 p.) : ill."]},{"key":"dc:title","label":"Title","values":["On-Device Semi-Supervised Learning For Healthcare Applications"]}]}],"canonical_facts":{"dc:contributor":["Biswas, Subir","Ren, Jian","Kulkarni, Sandeep","Traganitis, Panagiotis"],"dc:creator":["Roy, Avirup"],"dc:date":["2026"],"dc:description":["Thesis (Ph.D.)--Michigan State University. Electrical and Computer Engineering - Doctor of Philosophy, 2026","Mobile health monitoring applications heavily rely on Embedded Cyber-Physical Systems (E-CPS) to provide real-time measurement and diagnostics insights into user health in the areas of physical activity, hydration level, various vital signs, and lifestyle in general. These systems demand not only high accuracy but also rapid system performance in order to ensure timely health interventions. Delays in processing can result in suboptimal outcomes, especially in critical health scenarios where immediate actions are required. To reduce operational latency and improve performance, intelligent E-CPS typically employ pre-trained supervised learning models to infer patterns based on real-time health data. However, the traditional approach to training these models presents challenges, such as limited access to labeled health data, difficulties in developing personalized learning models, and privacy concerns arising from the transmission of sensitive health data to external servers for training.This dissertation introduces a semi-supervised learning (SSL) framework to address such limitations of conventional supervised learning in mobile health monitoring applications. The proposed SSL framework leverages new unsupervised clustering and cluster labeling techniques to learn from sparsely labeled data points, making it particularly useful in scenarios where sufficient labeled data is unavailable. By training directly on data collected from each user’s device, this approach facilitates person-specific learning, enabling the system to adapt to the unique health patterns of individual users. The on-device learning mechanism ensures that all data remains on the user’s device, eliminating the need for data transmission to external servers, thereby enhancing privacy and reducing communication costs. Additionally, the framework is optimized for resource-constrained devices, consuming minimal computational power and battery life, making it ideal for wearable and mobile challenges of real-time health monitoring while ensuring that devices remain responsive and effective for continuous, personalized health tracking.","Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).","Includes bibliographical references."],"dc:format":["application/pdf","1 online resource (146 p.) : ill."],"dc:identifier":["etd:52950","local:Roy_grad.msu_0128D_21961","https://doi.org/doi:10.25335/kc65-vd80"],"dc:language":["eng"],"dc:relation":["Electronic Theses & Dissertations"],"dc:rights":["Attribution 4.0 International"],"dc:subject":["Electrical engineering","Computer engineering"],"dc:title":["On-Device Semi-Supervised Learning For Healthcare Applications"],"dc:type":["text","Electronic dissertations"]},"updated_at":"2026-11-18T18:59:34Z"}